[PostgreSQL] B-Tree Deduplication With A Catch
By now everyone would know that PostgreSQL 13 is the best version available for anyone. One of the new features that worth mentioning is the way b-tree index is stored for duplicate values. If you have columns with low-cardinality and have duplicate values in them, Postgres will now store them only once. This means saving space and easier on the I/O for searching the values. Because the performance gain during read is very real and so b-tree deduplication is very recommended.